NER models for personal data anonymisation in public administration texts. 8 language/domain combinations. MAPA project, by Pangeanic.
AI & ML interests
None defined yet.
Organization Card
Pangeanic
Pangeanic develops multilingual AI technologies and developer tools for trusted data pipelines, model evaluation, language processing, and sovereign AI deployment.
Our expertise spans:
- Multilingual and multimodal data collection
- Speech and video datasets
- Data annotation and human evaluation
- Model alignment and benchmarking
- Machine Translation (MT) and Machine Translation Quality Estimation (MTQE)
- Document AI and intelligent document processing
- Data anonymization and privacy-preserving AI
- Terminology and glossary management
- Secure enterprise AI infrastructure
We publish models, datasets, tools, APIs, integrations, and technical resources that enable developers, researchers, and organizations to build reliable, production-ready multilingual AI systems.
For more information, visit https://www.pangeanic.com/.
models 8
Pangeanic/mapa-mt-administrative
Token Classification • Updated • 9
Pangeanic/mapa-ga-legal
Token Classification • Updated • 5
Pangeanic/mapa-lv-legal
Token Classification • Updated • 5
Pangeanic/mapa-fr-medical
Token Classification • Updated • 10
Pangeanic/mapa-de-legal
Token Classification • Updated • 7
Pangeanic/mapa-en-administrative
Token Classification • Updated • 12
Pangeanic/mapa-multilingual-administrative
Token Classification • Updated • 17
Pangeanic/mapa-es-legal
Token Classification • Updated • 16